Abstract:
Input to the Most Navigable Path (MNP) problem consists of the following: (a) a road network represented as a directed graph, where each edge is associated with numeric attributes
of cost and “navigability score” values; (b) a source and a destination and; (c) a budget value
which denotes the maximum permissible cost of the solution. Given the input, MNP aims to
determine a path between the source and the destination which maximizes the navigability
score while constraining its cost to be within the given budget value. The problem can be
modeled as the arc orienteering problem which is known to be NP-hard. The current stateof-the-art for this problem may generate paths having loops, and its adaptation for MNP
that yields simple paths, was found to be inefficient. In this paper, we propose five novel
algorithms for the MNP problem. Our algorithms first compute a seed path from the source
to the destination, and then modify the seed path to improve its navigability. We explore
two approaches to compute the seed path. For modification of the seed path, we explore different Dynamic Programming based approaches. We also propose an indexing structure for
the MNP problem which helps in reducing the running time of some of our algorithms. Our
experimental results indicate that the proposed solutions yield comparable or better solutions while being orders of magnitude faster than the current state-of-the-art for large real
road networks.
